Mechanisms by which thin filament variants induce hypertrophic cardiomyopathy
Hypertrophic cardiomyopathy (HCM) is a complex and highly heterogeneous disease that can result in asymmetric cardiac hypertrophy, myofibrillar disarray, and serious arrhythmogenesis that can cause sudden cardiac arrest (SCA). In fact, HCM is the leading cause of SCA in the youth, and elite athletes. HCM is a genetic disorder inherited in an autosomal dominant pattern with a frequency of about ~1/500. To date over 1,400 HCM-associated variants have been identified > 12 genes encoding sarcomeric proteins. Variants in genes encoding thick filament proteins cardiac myosin ß heavy chain (MYH7) and myosin binding protein C (MYBPC3) account for ~40% of HCM cases. On the other hand, genes encoding for thin filament proteins troponin I (TNNI3), troponin C (TNNC1), and troponin T (TNNT2) account for ~30% of HCM cases. These are of particular interest, however, because while the degree of hypertrophy could be minimal, the arrhythmogenicity could be serious enough to result in SCA. Thus, clinical screening of patients using echocardiography will likely result in a false negative diagnosis, yet they remain at high risk for SCA and is the focus of this proposal.

The overall objective of this proposal is to understand better how thin filament variants associated with HCM can impact cardiac contraction and arrhythmogenesis and strategies for effective therapies. The focus will on using human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) harboring variants in the genes encoding TNNC1, TNNI3, and TNNT2. We will use a suite of complementary approaches to delve deeper into the mechanisms of pathogenicity.

Over the last few years, exciting progress has been made in certain aspects of HCM and has resulted in the very recent FDA approval (April 2022) of the first and only drug for its treatment, mavacamten, a myosin ATPase inhibitor. A very recent focus has been on the myosin state during diastole, including a super-relaxed (SRX or closed) and a disordered relaxed (DRX or open) state. In the healthy heart, 40–50% of the myosin heads are in the SRX state with negligible ATP consumption, whereas in HCM there is a shift in this ratio, with only 15–20% of the myosin heads being in the SRX state. The increased myosin in the DRX state with HCM increases ATP consumption and the myosin heads are more primed to interact with actin. This cause excess myosin–actin cross-bridges during both systole and diastole leading to a hypercontractile state and diastolic dysfunction. Targeting the myosin ATPase with mavacamten has been shown to increase myosin SRX conformations by acting on multiple stages of the myosin chemomechanical cycle.

However, as promising as this seems, there is now growing evidence that mavacamten is not as effective in treating HCM cases resulting from troponin variants, despite the phenotype for this subclass also exhibiting a hypercontractile state and impaired relaxation. Potential mechanisms for thin filament associated HCM phenotype include alterations of the tropomyosin positioning on actin and increased myofilament Ca2+ sensitivity.

Thus, the aim of this proposal is to study the mechanisms of the HCM-related pathogenicity of TNNC1, TNNI3, and TNNT2 variants in terms of the contractile and electrophysiological state using state of the art models including reconstituted thin recombinant proteins, hiPSC-CMs organized in both 2D (monolayers) and 3D (cardioids) tissue structures.

Using hiPSC-derived atrial tissue to understand better the role of SK ion channel variants in atrial fibrillation
SK channels (SK1, SK2, and SK3) are Ca-activated K channels encoded by three genes, KCNN1, KCNN2, and KCNN3. While the tissue specificity and function of these ion channels remain controversial, there is a strong Genome-Wide Association Studies (GWAS) associated correlation, for example, with SK3 ion channel variants (e.g., 1q21 rs13376333 (RS)) and atrial fibrillation (AF). While GWAS play an important role in hypothesis generation, the data are only correlative in nature and not proof of causality. In a series of elegant studies, our collaborator L Hove Madsen used patient-derived native atrial appendage cells to study the electrophysiological properties, live cell imaging, and protein distribution that are distinct in AF patients. Although sophisticated, these studies do have shortcomings common to all native human heart tissue studies including co-morbidities, chronic drug treatment, and lack of suitable controls.

We used CRISPR Cas9 to create the AF-associated RS variant along with their isogenic controls (WT) in hiPSCs. This proposal includes some exciting preliminary data from these variant harboring hiPSC-derived atrial cardiomyocytes (hiPSC-aCMs). The same genome editing strategies /constructs will be used to create these in additional hiPSC lines that were generated from both male and female healthy donors. We also have adopted a hiPSC-CM maturation protocol which significantly enhances mitochondrial density, metabolic capacity, the formation of T-tubules and dyads, and myofibrillar alignment.

Arrhythmogenicity will be tested on 3D bioprinted atrial tissue (3DBAT) over a range of stimulation frequencies (1-4 Hz), ± ß adrenergic agonists, and rotor-inducing stimulation protocols. Using high-speed (1000 fps) optical mapping (OM), action potential duration (APD), Ca transient duration, and conduction velocity and dispersion will be measured. The potential for early afterdepolarizations, re-entry loops, and rotors will be determined. Once the OM is complete, Cell Painting will be used to examine cell nuclei, sarcomere alignment, collagen formation, and the cell-to-cell arrangement. Our aims are to use:
1. 3DBAT made of hiPSC-aCMs (± RS) and a bioink for the ECM, to test the hypothesis that APD prolongation will be observed in RS and an increased potential for arrhythmogenicity.
2. 3DBAT made of hiPSC-derived atrial cells and fibroblasts. Both cell types will be made from the same hiPSC line (± RS) at a ratio of 4:1 to test the hypothesis that co-bioprinting with fibroblasts will mature the hiPSC-aCMs based on their proteomic and morphological profile.
3. same 3DBAT as in aim 2 except TGFß will be used to activate fibroblasts to collagen-producing myofibroblasts, to test the hypothesis that atrial fibrosis will change the biomechanical and electrical properties of the atrial tissue increasing the arrhythmogenicity.
4. bioreactor based 3D differentiation of hiPSCs to atrial cardiomyocytes. While the other three aims are not predicated on the success of this aim, achieving this will be transformative for the field by generating hundreds of millions of hiPSCaCM all exposed to identical conditions.

This approach to understanding the mechanisms of familial AF exploits a cutting-edge platform of hiPSC-derived atrial tissue and an impressive array of phenotyping tools to understand the mechanisms of atrial arrhythmogenicity. The data generated should offer deeper insight into potential personalized therapeutic AF interventions.

Modeling human heart morphogenesis with bioreactor-derived cardiac organoids
Congenital heart disease (CHD) is the most common organ malformation in newborns. A genetic etiology can be linked to about 50% of CHD. Developing therapeutic interventions based on this genetic information requires understanding how gene mutations cause CHD. Experiments in model organisms have provided invaluable insights, but human heart development has unique features that are not replicated in animal models, particularly commonly used models such as mice and zebrafish. For example, cardiac gene expression differs in humans, making animal models unable to mimic certain aspects of human development and disease. Clearly a model of human heart development would be invaluable for further mechanistic studies and for the development of therapeutic interventions based on our growing understanding of genetic causes and pathogenic mechanisms. We developed a protocol to reproducibly generate cardiac organoids (COs) from human induced pluripotent stem cells (iPSCs) in 15 days. The protocol uses a stirred bioreactor, which produces controlled hemodynamic forces and carefully regulates the cell environment. Our bioreactor-derived COs (bCOs) undergo self-patterning and morphogenesis, forming hollow spheres several fold larger than previously described COs. The walls, primarily composed of iPSC-CMs, envelop central cavities reminiscent of cardiac chambers. We found that COs also contain endothelial cells and fibroblasts, and mutant COs lacking a cardiac specific RNA binding protein exhibited morphological defects. Based on these observations, we propose to establish bCOs as a novel platform for investigation of CHDs in the human context.